Conversations in Cardiology: Post-PCI Clopidogrel in CHAMPION PHOENIX Trial of Cangrelor
Morton Kern, MD, of the University of California, Irvine, often engages his colleagues via email in brief, informal dialogue on clinically relevant topics in interventional cardiology. With permission from the participants, TCTMD presents their conversations for the benefit of the interventional cardiology community. Your feedback is welcome—feel free to comment at the bottom of the page.
Kreton Mavromatis, MD, of Emory University School of Medicine and Atlanta VA Medical Center (Atlanta, GA), asks:
Public Citizen has accused the CHAMPION PHOENIX trial of being unethical due to the high incidence of clopidogrel loading after PCI, and have asked for an investigation into how widespread the practice of post-PCI clopidogrel loading is.
While many of our ACS patients are on clopidogrel immediately prior to PCI, many of our stable CAD patients may not be. What is your timing of clopidogrel/other P2Y12 inhibitor loading in patients undergoing planned elective PCI or diagnostic coronary angiography with ad hoc PCI who are not already on P2Y12 inhibitor for other clinical reasons?
This attack seems like a lawsuit in process, and the documents appear court-ready. I don’t know anything about Public Citizen MDs or their motivation. Maybe the competing drug interests are somewhere in the background.
Nonetheless, our practice is as commonly talked about: give clopidogrel before the procedure when possible and within 2 hours after the procedure if not possible. Generally we don’t wait 6 hours, and I can’t remember a case but would not be surprised if a post-PCI order was not picked up until later with drug administration delayed by 6 hours.
While data suggest best outcomes occur with early administration, we rarely see adverse events for post-procedure drug routines. Keep up informed as this review proceeds.
Jeffrey W. Moses, MD, of Columbia University Medical Center (New York, NY), replies:
Sid Wolfe is a gadfly. The Pharma Ralph Nader. Highly political for decades. As for elective loading, the latest meta-analyses do not indicate any advantage and practices vary widely given the surgical issues that dominate certain centers.
This is an incredibly important question, as our surgeons adhere to a “clopidogrel washout” prior to CABG in the event of multivessel disease requiring more than stents [that amounts to] in general a 3- to 5-day delay in surgery. I admittedly operate in a “data-free” zone in patients naive to P2Y12 inhibitors: radial approach, unfractionated heparin, tirofiban with higher-risk PCI, load with whatever P2Y12 antagonist is appropriate after the procedure, stop the tirofiban when the bag runs out. I try to avoid P2Y12 inhibitors for elective or lower-risk NSTEMI cases until I know the anatomy. We don’t have a consistent approach in our lab with respect to anticoagulation, but most of us appreciate the pain of the clopidogrel washout in elective cases for our surgeons.
We record this time variable and try to keep it within 30 minutes of PCI as a performance measure. These drugs are kept in the cath lab and, if the operator feels strongly, ticagrelor oral slurry (crushed) is often given in the lab on the table for ACS patients who have not been pretreated. Stable patients are usually given oral clopidogrel 600 mg within 30 minutes of stent deployment immediately upon arriving to the CV recovery unit based on preprinted protocol-driven post-PCI orders. We review cases that fall out of the 30- to max 45-minute window.
Yes, we experienced this accusation at the FDA cangrelor panel presentation, which is at best uninformed and at worst intentionally malicious (for reasons not known to me). The latest data I have seen regarding US practice patterns on preloading are from a US-based cross-sectional study published in 2010 showing that first use of clopidogrel in patients undergoing PCI is approximately equally distributed before, during, and after the cath lab. So only one-third of patients were actually preloaded. Does anyone know of any more recent data?
Moreover, there are no worthwhile data [showing] that preloading in stable CAD is beneficial (in fact, most studies suggest it is harmful, with increased major bleeding), and by intention to treat preloading has also not been clearly proven to be beneficial in NSTE-ACS, while again increasing bleeding. While there is some evidence that having NSTE-ACS patients preloaded is helpful if PCI is performed, we rarely have the luxury of knowing who those patients are before angiography (it’s only about 65%), and the others have increased bleeding and/or delayed surgery. The data for preloading in STEMI are promising but weaker, and none of the oral agents are effective in the first hour in STEMI when they are really necessary, during the primary PCI procedure. In fact, the only agent that has proven efficacy for preloading is cangrelor—a rapidly acting IV agent which is given once the coronary anatomy is known (ie, PCI will be done), is maximally alive within minutes, and is more potent than any of the oral agents (~95% inhibition of ADP-induced platelet aggregation). In patients who are P2Y12 inhibitor naive, cangrelor decreases 48-hour ischemic MACE by ~20% and stent thrombosis by ~40%, while increasing mild but not severe bleeding. Of note, cangrelor had similar relative efficacy in patients with stable CAD, NSTE-ACS and acute/recent STEMI, and as such offers great flexibility for a wide range of patients in the cath lab.
These are the data, among others, that Gregg is talking about. The data supporting preloading are weak for stable and NSTEMI. We all usually complete the PCI before the oral agent has reached therapeutic effect with an oral agent in STEMI (or there are bigger problems with our STEMI system of care than preloading). At best the issue of preloading is controversial and as such “unethical” is a strong/inaccurate term to use. As for timing of the load after the stent is deployed, it stands to reason that one would want the time period to be as short as possible where the patient has a new stent and does not have adequate anti platelet effect.
Ajay J. Kirtane, MD, SM, of NewYork-Presbyterian Hospital/Columbia University Medical Center (New York, NY), replies:
At Columbia our practice for many years has been to preload elective and otherwise uncomplicated patients with 600 mg of clopidogrel 2 hours before the procedure (the rationale being that, one, these patients are low risk of having to go to surgery urgently/emergently, and two, several of our surgeons will operate on clopidogrel). When ticagrelor became available, our protocol for NSTE-ACS patients changed based upon the PLATO results to load with 180 mg ticagrelor on admission, but in all honesty because of the surgical considerations this often did not happen. Interestingly, this was the case even though the delay with ticagrelor to surgery is less than that with clopidogrel (see a recent European Heart Journal analysis from Sweden for more than just pharmacodynamic confirmation of this). But because our surgeons have been more reluctant to operate on ticagrelor, we still see a great deal of variability in practice.
In part now because of the flexibility an IV P2Y12 inhibitor affords, we will soon be moving away from routine preloading of NSTE-ACS patients who will go to the cath lab quickly. For those with longer times to the lab, we will likely still preload with oral agents. We will likely keep our current practice of loading straightforward electives due to cost considerations and will continue loading STEMI patients with more potent oral P2Y12 agents because so few go to CABG.
I do think that some of the criticism of CHAMPION PHOENIX stems from the belief that the comparison was an “unfair” one and this is what led to the positive study results. While the positive study results may indeed have reflected the fact that the control arm likely didn’t have inhibited platelets at the time of the PCI, isn’t that precisely why we need an IV agent? And it is clear based upon “community-practice” standards that the majority of patients in the United States do not have PCI while adequately preloaded, almost entirely due to the potential CABG consideration.
This is yet another area where the dissociation between idealized medicine and real-world clinical medicine is clearly manifest.
Samuel M. Butman MD, MHA, Heart & Vascular Center of Northern Arizona (Cottonwood, AZ), replies:
Here and the University where I worked for many years, our surgeons (strongly) preferred we did not send them an occasional patient who had been preloaded. We here and at the U acquiesced to their wishes.
In my experience overseeing my practice and that of others around me, I cannot recall a case where an acute stent thrombosis, fortunately rare of course these days, could not be explained by some other oversight, mishap, clinical, or anatomical issue.
My colleagues and I continue with our practice of loading immediately in the cath lab (unless the patient is intubated, of course, which sets up a whole bunch of untested solutions). That has been and continues to be a win-win.
I think there are 2 points here:
1. Wolfe 's complaint about the CHAMPION trial is not a valid issue in that there is no community standard for elective PCI. He … has gone after numerous drugs/pharma companies and to my knowledge has never been successful.
2. There is no community standard, because there are no compelling data in that clinical scenario.
We instituted loading in our practice as a carryover from CREDO and other data supporting that 600 mg had a better effect than 300 mg on platelet parameters well over a decade ago. The subsequent clinical data haven’t supported that, but we maintain it as it seems to have little downside with our low urgent CABG rates in electives and it keeps things simple for the nurses and physician assistants. But I also practice at a high-volume site where elective loading is not routine and am comfortable with when they forget to preload.
David E. Kandzari, MD, of Piedmont Heart Institute (Atlanta, GA), replies:
Coming a bit late into the conversation, I participated in the second cangrelor panel, and the 3 opponents who revisited the preloading ‘ethics’ were by my recall 2 statisticians and 1 nephrologist. There were only 2 interventional cardiologists on the panel—one whose center does not preload (Piedmont, me) and one that routinely preloads (Johns Hopkins).
Specific to cangrelor, the physician presenting on behalf of cangrelor stated in his slides that he would consider the drug for any patient undergoing PCI who did not receive pretreatment. That certainly will not represent our practice, but I do (and did at the panel) believe there specific instances in which the drug has a unique benefit. For elective PCI, we routinely load at the time of PCI with prasugrel given more immediate efficacy, then transition to clopidogrel, keep prasugrel, transition to ticagrelor. For emergency PCI, patients are loaded either with prasugrel or ticagrelor pending eligibility.
One of our institution's greatest financial challenges is hospital stay among patients on DAPT who require CABG. Our surgeons perform platelet assays in an effort to abbreviate the delay, but overall refuse to operate on patients taking DAPT (unless recovery of antiplatelet activity). Upon hearing that some centers have surgeons who will operate on DAPT, they’ve asked me to present to them next month. So, if anyone has data that would support CABG on DAPT, please let me know.
Andrew Doorey, MD, of Christiana Care Cardiology Consultants (Newark, DE), replies:
I didn’t make the connection of Public Citizen and Sidney Wolfe [earlier]. But now I remember. He made a big splash in our town about 10 years ago when he led a campaign to have Crestor banned. Here is his petition, which got a lot of press.
Since Astra Zeneca is based here in Wilmington, this was a big deal in the local news. I was concerned because while we all know about rhabdomyolysis, I really wasn't aware of any reliable kidney failure data for rosuvastatin, which was just getting to be widely used. I grilled our local reps and went pretty far up the adverse-effect chain at Astra Zeneca trying to find out if there is any validity to this. I could find none. Then, as you may recall, the FDA did special postmarketing evaluation. A year or so later they came out with about a 75-page paper, and I was so interested in the results, I recall joking that I may have been the only one who read it. (By the way, I have no connection with AstraZeneca, and I’m not on their speakers bureau and don’t get any money or anything from them).
Long story short, all the accumulated data suggested that the Crestor actually improved renal function (which I thought was logical since atherosclerosis probably contributes to renal dysfunction in lots of ways). As far as I know, Public Citizen dropped their crusade against Crestor at about that time.
I remember searching the literature to find out any reliable reports of kidney failure with Crestor, as I recall found none. So this group basically made stuff up and created a sizable national concern. They seemed to be more interested in the publicity than the fact.
The timing of P2Y12 inhibitor administration is usually within 2 to 6 hours post-PCI with a 600 mg oral dose of clopidogrel. However, there is no community standard, because there are no compelling data in that clinical scenario. Finally, as said by Gregg Stone, “this accusation at the FDA cangrelor panel presentation, is at best uninformed and at worst intentionally malicious.” If needed, such complaints can be addressed by the PI’s of the study.